Titanium zeolites, i.e., synthetic molecular sieves that incorporate titanium atoms in a silicate framework, catalyze a wide variety of valuable oxidative organic reactions. The versatility of titanium zeolites, particularly TS-1, for arene hydroxylation, alkane oxidation, olefin epoxidation, thioether oxidation, Baeyer-Villiger oxidation reactions, and other important transformations is well known. For a review, see P. Kumar et al., Synlett. (1995) 289. Titanium zeolites catalyze the epoxidation of propylene with hydrogen peroxide. The hydrogen peroxide can be supplied (see, e.g., U.S. Pat. No. 6,037,484) or it can be generated “in situ” by using titanium zeolites that incorporate a transition metal, especially palladium (see U.S. Pat. No. 6,008,388 and references cited therein).
Unwanted hydrogenation of propylene to propane complicates the “in situ” epoxidation of propylene using hydrogen, oxygen, and a transition metal-containing titanium zeolite. Nitrogen compounds such as ammonium hydroxide or ammonium bromide have been added to suppress propane formation (see U.S. Pat. No. 6,008,388). Nitrogen-containing polymers were not suggested.
Another side reaction is common to olefin epoxidations with hydrogen peroxide: zeolite-promoted ring opening of the epoxide with water and/or alcohol solvents to give glycols and glycol ethers. For example, when PO is made from propylene using TS-1 as a catalyst in aqueous methanol, ring opening reactions typically limit the PO/POE (molar ratio of propylene oxide to propylene oxide plus ring-opening products that derive from PO) to about 91% at 50° C. Selectivity deteriorates as temperature increases. In addition to sacrificing valuable epoxide product, ring opening introduces impurities that must be removed. U.S. Pat. No. 6,037,484 teaches to add 2,4-lutidine or another substituted pyridine compound to the hydrogen peroxide feed to suppress ring opening. Soluble pyridines such as these must be continually replenished in any continuous process for making an epoxide. Polymers containing pyridine moieties are not suggested.
Recently, we discovered that polymer-encapsulated titanium zeolites are valuable oxidation catalysts (see, e.g., copending application Ser. No. 10/796,842, filed Mar. 9, 2004), particularly for olefin epoxidations. In particular, we found that polymer encapsulation improves catalyst filterability (an advantage for both catalyst preparation and catalyst recovery) and provides a significant improvement in selectivity to propylene oxide (from 91% to 93% PO/POE). We also found that polymer-encapsulated transition metals effectively catalyze the reaction of hydrogen and oxygen to make hydrogen peroxide (see copending application Ser. No. 10/796,810, filed Mar. 9, 2004). Vinylpyridine polymers were not disclosed.
While the pharmaceutical industry has used polymer encapsulation for years to mask taste, impart storage stability, reduce stomach irritation, target delivery, or control release of drugs, benefits of the technique for catalysis are just now being realized (for examples, see Chem. Commun. (2003) 449 and references cited therein; Angew. Chem. Int. Ed. 40 (2001) 3469; J. Am. Chem. Soc. 120 (1998) 2985).
In sum, the industry would benefit from improved oxidation catalysts. In particular, the industry needs olefin epoxidation catalysts that provide good selectivity while minimizing ring-opening side reactions. Catalysts that can provide good selectivity over a wide temperature range would be especially valuable. Ideally, the catalysts would be inexpensive and easy to make. Catalysts for making hydrogen peroxide directly from hydrogen and oxygen are also needed.